Systemic juvenile rheumatoid arthritis (sJRA) is associated with significant long term morbidity and mortality. Current therapies including methotrexate are considered ameliorative rather than remission inducing or curative. There have been anecdotal reports suggesting that pulse therapy with intravenous corticosteroids, cyclophosphamide, and methotrexate may induce prolonged remissions in sJRA. Therefore, the objectives of this study are to determine and compare the ability of two pulse therapy regimens to induce remission in sJRA of less than 12 months duration during a 9-month, open-label, randomized, actively controlled clinical trial. The first pulse therapy regimen is composed of intravenous methylprednisalone 30 mg per kg per day (1 gram max) for three consecutive days, intravenous cyclophosphamide 0.4 grams per meter2 BSA on the third day, and up to 20 mg per meter2 per week of methotrexate. The second pulse therapy regimen is identical to the first, except no cyclophosphamide is given. Up to five cycles of these regimens may be given over a 9-month period. Patients in both groups may also receive background medications including a non-steroidal, anti-inflammatory drug and up to 0.5 mg/kg/d of oral prednizone. The primary outcome to measure the effect of this therapy will be the proportion of patients who achieve clinical remission according to the ACR criteria for remission in rheumatoid arthritis. Secondary outcome measures of effectiveness include proportion of patients who demonstrate clinical response according to the preliminary definition of improvement for JRA. In addition, time per remission and the duration of remission will be compared between the two groups among patients who do remit. Specific Aim 2 is to determine and compare the short and intermediate term (18 months) safety profiles of the pulse therapy regimens as defined in Specific Aim 1. Long term goals of the project are to determine and compare the longer-term safety profiles of this treatment regimen. This will involve analysis of patient/parent-derived data obtained by mail or phone follow-up to detect significant medical problems and reproductive or neoplastic complications.